The focus on AI safety disasters ignores everyday harms
TL;DR
Consider what passes for success in technology policy. An AI company promises that its chatbot will not encourage suicide. A social media platform makes it more difficult for children to buy fentanyl. These are important protections, and they should have existed all along.
Nauti's Take
A small team should start with the ordinary workday: Which bad recommendations, dependencies, or decisions appear when the system is used repeatedly? Before rollout, test those cases with real user data, clear escalation paths, and a rule for stopping the system.
That is the evidence needed to judge whether a safety promise holds in practice.
Sources